hide
Free keywords:
-
Abstract:
We present a high-resolution 3D model of the 3D structure and cellular composition of the rat thalamo-cortical vibrissal system. Despite the fact that the vibrissal system is widely used in neuroscience research, effects of its 3D structure on connectivity and function are often overlooked. A realistic anatomical model of network connectivity has to take these parameters into account.
To do so, we automatically reconstruct anatomical landmarks (barrels in cortical Layer 4, pial and white matter surfaces, and barreloids in vibrissal thalamus (VPM)) in 3D from high-resolution microscope images. Surprisingly, we find that the anatomical structures vary significantly across the barrel field. For example, the column volume increases three-fold between different rows. In contrast, the overall 3D layout remains remarkably preserved across animals. The position of individual barrels varies by only 35μm between different animals, and the orientation of individual columns by only 4.5°. This small variability allows creating a standardized, high-resolution 3D model of the layout of barrel cortex.
To determine whether the cellular composition of the vibrissal system is affected by these large geometric differences, we automatically detect all neuron somata in 3D confocal images of vibrissal cortex and thalamus. The number of neurons in a column varies from 9300 to 29000 across the vibrissal cortex. In contrast, the density of neurons and inhibitory interneurons is constant in all columns (81,000/mm^3 and 8,000/mm^3, respectively). The number of neurons in a barreloid varies from 90 to 400 across VPM. Despite these large variations, the number of cells in a cortical column correlates strongly with the number of cells in the corresponding barreloid in VPM. Further, because the number of neurons in individual columns is preserved between different animals, registration of individual 3D neuron distributions to the standardized 3D barrel cortex allows creating a standardized average 3D neuron distribution.
These results show that despite large differences between different columns and barreloids, the 3D layout and cellular composition of individual columns and barreloids is well-preserved between different animals. This allows creating a standardized 3D model of the geometry and cellular composition of the vibrissal system. Further, these results indicate that a cortical column may not be the elementary functional unit of mammalian cortices, as is commonly believed.
